Method and device for evaluating quality of service of data center, and storage medium

Abstract

The application discloses a data center service quality evaluation method and device, and a storage medium. It relates to the technical field of big data. The method comprises the following steps: obtaining score values of N target performance indexes of an object to be evaluated, wherein the N target performance indexes comprise primary performance indexes and S secondary performance indexes; determining a cloud model based on a preset grading standard and the score values of each target performance index; determining the weight of each secondary performance index based on a target model, wherein the target model is a neural network model; determining the membership degree of the object to be evaluated belonging to each service quality level based on the cloud model and the weight of each secondary performance index, and determining the evaluation result of the object to be evaluated based on the membership degree of the object to be evaluated belonging to each service quality level. The application solves the technical problem of poor evaluation effect in the related art that the total evaluation index is used as the framework for quantitative evaluation of the service quality of the data center.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data, in particular to a data center service quality evaluation method and device and storage medium. BACKGROUND

[0002] Since the data center belongs to the high energy consumption industry, the development of the data center is mainly focused on the reduction of energy consumption, thereby ignoring the improvement of the service quality provided by the data center. In order to manage and improve the service quality of the data center, the service quality of the data center needs to be evaluated. However, the method for evaluating the service quality of the data center in the related art mainly has the following methods:

[0003] There are two problems in the evaluation of the service quality of the data center:

[0004] 1. The weights of various indexes are manually set uniformly, and then various data processing methods are used to fuse multiple indexes, which ignores the differences between the object types of the services of various data centers, and the dynamic and changeable behaviors of the clients, and lacks a method for determining the weights based on user evaluation, so it is difficult to adapt to various types of data centers.

[0005] 2. Simply taking the total index as the framework and using a quantitative method, the complexity of the data center itself is ignored, which leads to the deviation of the evaluation of the service quality of the data center from the characteristics of the data center itself, and it is difficult to accurately evaluate the service quality of the data center.

[0006] In view of the above problems, no effective solution has been proposed. SUMMARY

[0007] The embodiments of the present application provide a data center service quality evaluation method and device and storage medium, which at least solve the technical problem that the total evaluation index is uniformly taken as the framework in the related art, and the service quality of the data center is quantitatively evaluated, and the evaluation effect is poor.

[0008] According to an aspect of the embodiments of the present application, there is provided a method for evaluating service quality of a data center, comprising: obtaining score values of N target performance indicators of an object to be evaluated, wherein the object to be evaluated comprises a data center of a service quality level to be evaluated, the N target performance indicators comprise a primary performance indicator and S secondary performance indicators, the primary performance indicator is composed of the S secondary performance indicators, N and S are positive integers greater than 1, and S is less than N; determining a cloud model based on a preset grading standard and the score values of each target performance indicator, wherein the preset grading standard is used to evaluate the service quality level of the object to be evaluated; determining a weight of each secondary performance indicator based on a target model, wherein the target model is a neural network model obtained by training an initial neural network model based on indicator data in a target time period; determining a membership degree of the object to be evaluated belonging to each service quality level based on the cloud model and the weight of each secondary performance indicator, and determining an evaluation result of the object to be evaluated based on the membership degree of the object to be evaluated belonging to each service quality level, wherein the evaluation result is used to indicate the service quality level of the object to be evaluated.

[0009] Further, the step of determining the cloud model based on the preset grading standard and the score values of each target performance indicator comprises: determining a boundary value corresponding to each service quality level based on the preset grading standard, wherein the boundary value comprises a minimum value and a maximum value of a score value range associated with each service quality level; determining a digital feature of the cloud model based on the boundary value corresponding to each service quality level, wherein the digital feature comprises an expectation, an entropy, and a hyper entropy; and determining the cloud model based on the digital feature.

[0010] Further, each secondary performance indicator is composed of T tertiary performance indicators, T is a positive integer, and the target model is obtained by: obtaining the indicator data, wherein the indicator data comprises indicator values of all tertiary performance indicators in the target time period and score values of the primary performance indicator in the target time period; and training an initial neural network model based on the secondary performance indicators, the indicator values of all tertiary performance indicators in the target time period, and the score values of the primary performance indicator in the target time period to obtain the target model.

[0011] Further, the step of training the initial neural network model to obtain the target model based on the secondary performance indicators, the indicator values of all the tertiary performance indicators in the target time period and the score values of the primary performance indicators in the target time period comprises: determining the activation function of the hidden layer of the initial neural network model based on the S secondary performance indicators; taking the indicator values of all the tertiary performance indicators in the target time period as the input of the input layer of the initial neural network, taking the score values of the primary performance indicators as the training labels, and training the initial neural network model based on the activation function to obtain the target model.

[0012] Further, the step of determining the weight of each secondary performance indicator based on the target model comprises: obtaining the weight coefficient associated with the target model; and determining the weight of each secondary performance indicator based on the weight coefficient.

[0013] Further, the step of determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each service quality level comprises: determining the maximum membership in the membership of the to-be-evaluated object belonging to each service quality level; and determining the evaluation result based on the service quality level associated with the maximum membership.

[0014] Further, after determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each service quality level, the method further comprises: generating a cloud chart associated with the cloud model, wherein the cloud chart is used to at least display the mean value of each target performance indicator; and displaying the cloud chart.

[0015] Further, after determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each service quality level, the method further comprises: determining a first sequence based on the membership of the to-be-evaluated object belonging to each service quality level, wherein the type of the first sequence comprises a mother sequence associated with a grey correlation analysis method; determining a second sequence based on the N target performance indicators, wherein the type of the second sequence comprises a characteristic sequence associated with the grey correlation analysis method; and determining the correlation degree between each target performance indicator and the evaluation result by the grey correlation analysis method based on the first sequence and the second sequence.

[0016] According to another aspect of the embodiments of the present application, there is also provided an apparatus for evaluating service quality of a data center, comprising: a first obtaining unit configured to obtain score values of N target performance indicators of an object to be evaluated, wherein the object to be evaluated comprises a data center of which service quality level is to be evaluated, and the N target performance indicators comprise a primary performance indicator and S secondary performance indicators, the primary performance indicator is composed of the S secondary performance indicators, N and S are positive integers greater than 1, and S is less than N; a first determining unit configured to determine a cloud model based on a preset grading standard and the score values of the target performance indicators, wherein the preset grading standard is used to evaluate the service quality level of the object to be evaluated; a second determining unit configured to determine weights of the secondary performance indicators based on a target model, wherein the target model is a neural network model obtained by training an initial neural network model based on indicator data in a target time period; and a third determining unit configured to determine membership degrees of the object to be evaluated belonging to each service quality level based on the cloud model and the weights of the secondary performance indicators, and determine an evaluation result of the object to be evaluated based on the membership degrees of the object to be evaluated belonging to each service quality level, wherein the evaluation result is used to indicate the service quality level of the object to be evaluated.

[0017] Further, the first determining unit comprises: a first determining sub-unit configured to determine boundary values corresponding to each service quality level based on the preset grading standard, wherein the boundary values comprise minimum values and maximum values of score value ranges associated with each service quality level; a second determining sub-unit configured to determine digital features of the cloud model based on the boundary values corresponding to each service quality level, wherein the digital features comprise expectation, entropy and hyper entropy; and a third determining sub-unit configured to determine the cloud model based on the digital features.

[0018] Further, each secondary performance indicator is composed of T tertiary performance indicators, T is a positive integer, and the target model is obtained by the following units: a second obtaining unit configured to obtain the indicator data, wherein the indicator data comprises indicator values of all tertiary performance indicators in the target time period and score values of the primary performance indicator in the target time period; and a training unit configured to train an initial neural network model based on the secondary performance indicators, the indicator values of all tertiary performance indicators in the target time period and the score values of the primary performance indicator in the target time period, to obtain the target model.

[0019] Further, the training unit comprises a fourth determination subunit configured to determine an activation function of the initial neural network model hidden layer based on the S second performance indicators; and a training subunit configured to take all third performance indicator values in the target time period as input of the initial neural network input layer, take the score value of the first performance indicator as a training label, train the initial neural network model based on the activation function, and obtain the target model.

[0020] Further, the second determination unit comprises an acquisition subunit configured to acquire a weight coefficient associated with the target model; and a fifth determination subunit configured to determine a weight of each second performance indicator based on the weight coefficient.

[0021] Further, the third determination unit comprises a sixth determination subunit configured to determine a maximum membership degree of the to-be-evaluated object belonging to each service quality level; and a seventh determination subunit configured to determine the evaluation result based on a service quality level associated with the maximum membership degree.

[0022] Further, the data center service quality evaluation apparatus further comprises a generation unit configured to generate a cloud chart associated with the cloud model after determining the evaluation result of the to-be-evaluated object based on the membership degrees of the to-be-evaluated object belonging to each service quality level, wherein the cloud chart is used to at least display a mean value of each target performance indicator; and a display unit configured to display the cloud chart.

[0023] Further, the data center service quality evaluation apparatus further comprises a fourth determination unit configured to determine the evaluation result of the to-be-evaluated object based on the membership degrees of the to-be-evaluated object belonging to each service quality level, wherein the method further comprises a fifth determination unit configured to determine a first sequence based on the membership degrees of the to-be-evaluated object belonging to each service quality level, wherein the type of the first sequence comprises a mother sequence associated with a grey correlation analysis method; a sixth determination unit configured to determine a second sequence based on N target performance indicators, wherein the type of the second sequence comprises a characteristic sequence associated with the grey correlation analysis method; and a seventh determination unit configured to determine a correlation degree between each target performance indicator and the evaluation result based on the first sequence and the second sequence by using the grey correlation analysis method.

[0024] According to another aspect of the embodiments of the present application, an electronic device is also provided, which comprises a processor and a memory for storing executable instructions of the processor, wherein the processor is configured to execute the data center service quality evaluation method of any one of the above aspects by executing the executable instructions.

[0025] According to another aspect of the embodiments of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. When the computer program is executed, the computer readable storage medium controls a device where the computer readable storage medium is located to perform the data center service quality evaluation method according to any one of the preceding aspects.

[0026] In the present application, score values of N target performance indicators of an evaluation object are obtained, wherein the evaluation object includes a data center of an evaluation service quality level, and the N target performance indicators include a first-level performance indicator and S second-level performance indicators, the first-level performance indicator is composed of the S second-level performance indicators, N and S are positive integers greater than 1, and S is less than N; a cloud model is determined based on a preset grading standard and the score values of each target performance indicator, wherein the preset grading standard is used to evaluate the service quality level of the evaluation object; weights of each second-level performance indicator are determined based on the target model, wherein the target model is a neural network model obtained by training an initial neural network model based on index data in a target time period; the membership degrees of the evaluation object belonging to each service quality level are determined based on the cloud model and the weights of each second-level performance indicator, and the evaluation result of the evaluation object is determined based on the membership degrees of the evaluation object belonging to each service quality level, wherein the evaluation result is used to indicate the service quality level of the evaluation object. Thus, the technical problem of poor evaluation effect caused by the fact that the total evaluation indicator is used as the main indicator to quantitatively evaluate the data center service quality in the related art is solved. In the present application, the weights of the performance indicators are determined by the neural network model, and the service quality level of the evaluation object is determined by the cloud model based on the weights and the score values of the performance indicators. Thus, the situation that the quality evaluation result is not accurate caused by the fact that the characteristics of the data center are ignored and the weights of all indicators are uniformly used to evaluate the data center service quality in the related art is avoided. Thus, the technical effect of improving the accuracy of the data center service quality evaluation is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0027] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and serve to explain the principles of the application. In the drawings:

[0028] Figure 1 is a flowchart of an optional data center service quality evaluation method according to an embodiment of the present application;

[0029] Figure 2 is a schematic diagram of an optional performance indicator according to an embodiment of the present application;

[0030] Figure 3 is a schematic diagram of an optional neural network according to an embodiment of the present application;

[0031] Figure 4 is a schematic diagram of an optional data center service quality evaluation model according to an embodiment of the present application;

[0032] Figure 5 is a schematic diagram of an optional data center service quality evaluation device according to an embodiment of the present application;

[0033] Figure 6 is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the personnel in the art better understand the present application scheme, the technical scheme in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0035] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0036] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0037] Embodiment one

[0038] According to an embodiment of the present application, an optional method for evaluating the service quality of a data center is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0039] Figure 1 is a flowchart of an optional method for evaluating the service quality of a data center according to an embodiment of the present application, as shown in Figure 1 , the method comprises the following steps:

[0040] Step S101, obtaining the score value of N target performance indicators of the to-be-evaluated object, wherein the to-be-evaluated object includes a data center whose service quality level is to be evaluated, and the N target performance indicators include a primary performance indicator and S secondary performance indicators, the primary performance indicator is composed of the S secondary performance indicators, N and S are positive integers greater than 1, and S is less than N.

[0041] The to-be-evaluated object described above can be a data center whose service quality level is to be evaluated, and the N target performance indicators described above can include a primary performance indicator and S secondary performance indicators, each secondary performance indicator can further include a plurality of tertiary performance indicators.

[0042] Figure 2 is a schematic diagram of an optional performance indicator according to an embodiment of the present application, and the performance indicator in the present embodiment will be described below. Figure 2 , four types of sub-indicators (corresponding to the S secondary performance indicators) can be selected as the composition of the service quality total indicator (corresponding to the primary performance indicator, such as the data center service quality Q in Figure 2 , as shown in Figure 2 , the four types of sub-indicators are: availability (Q1), throughput (Q2), latency (Q3), and benefit (Q3), and in each type of sub-indicator, a plurality of operation and maintenance real-time data can be selected as feature indicators (corresponding to the tertiary performance indicators), the feature indicators of the availability are the number of performance monitoring interruptions, the number of rescheduling interruptions, and the number of non-maskable interruptions, the feature indicators of the throughput are the total number of transmitted data packets, the total number of accepted data packets, and the number of page cache memories, the feature indicators of the latency are the network card byte transmission rate, the network card byte reception rate, and the system load rate, and the feature indicators of the benefit are the CPU (central processing unit) utilization rate and the memory utilization rate.

[0043] Step S102, determining a cloud model based on the preset grading standard and the score value of each target performance indicator, wherein the preset grading standard is used to evaluate the service quality level of the to-be-evaluated object.

[0044] The preset grading standard can be used to evaluate the service quality level of the to-be-evaluated object. In an optional manner, in the preset grading standard, the service quality of the data center can be divided into five levels, i.e., top service (level I), first-class service (level II), good service (level III), poor service (level IV), and inferior service (level V), and each level can correspond to a score range.

[0045] In this embodiment, based on the boundary values of the score range of each service quality level in the preset grading standard, the numerical features of the cloud model, i.e., expectation (E x ), entropy (E n ), and hyper-entropy (H e ) can be calculated to obtain the cloud model.

[0046] In step S103, weights of each secondary performance indicator are determined based on the target model, wherein the target model is a neural network model obtained by training an initial neural network model based on indicator data in a target time period.

[0047] The target model can be a neural network model obtained by training an initial neural network model based on indicator data in a target time period. The target time period can be a historical time period. The indicator data can be indicator data generated by the to-be-evaluated object in the target time period, or can be indicator data generated by a data center of the same type as the to-be-evaluated object. The initial neural network model can be trained based on the indicator data to obtain a neural network model, and the weight coefficients in the trained neural network model can be extracted to obtain the weights of each secondary performance indicator.

[0048] In step S104, the membership degrees of the to-be-evaluated object to each service quality level are determined based on the cloud model and the weights of each secondary performance indicator, and the evaluation result of the to-be-evaluated object is determined based on the membership degrees of the to-be-evaluated object to each service quality level, wherein the evaluation result is used to indicate the service quality level of the to-be-evaluated object.

[0049] In this embodiment, the membership degrees of the sample data (target performance indicators) to the service quality levels are calculated, and then the service quality rating cloud diagrams corresponding to the performance indicators are obtained in combination with the rating standards corresponding to the preset grading standards. When calculating the membership degrees, the weights w=(w1, w2, w3, w4) of each secondary performance indicator can be used. The sample data is substituted into the cloud model, and the sample data and the weights are comprehensively calculated to obtain the membership degrees of the to-be-evaluated object to each service quality level. The maximum membership degree max{U p,k} (the membership degree of the sample p to the level k) corresponds to the corresponding energy efficiency level (corresponding to the quality service level), and then the evaluation result is obtained.

[0050] Through the above steps, in this embodiment, the weight of the performance index is determined through the neural network model, and based on the weight and the score value of the performance index, the service quality level of the to-be-evaluated object is determined through the cloud model, which avoids the case that the related art ignores the characteristics of the data center itself, uniformly weights each index to evaluate the service quality of the data center, and the quality evaluation result is not accurate, thereby realizing the technical effect of improving the accuracy of the data center service quality evaluation. Further, the technical problem of the related art of uniformly taking the total evaluation index as the framework to quantitatively evaluate the data center service quality and poor evaluation effect is solved.

[0051] Optionally, based on the preset grading standard and the score value of each target performance index, the step of determining the cloud model comprises: determining the boundary value corresponding to each service quality level based on the preset grading standard, wherein the boundary value comprises: the minimum value and the maximum value belonging to the score value range associated with each service quality level; determining the digital features of the cloud model based on the boundary value corresponding to each service quality level, wherein the digital features comprise: expectation, entropy, and hyper entropy; and determining the cloud model based on the digital features.

[0052] In this embodiment, a corresponding preset grading standard L = {I, II, III, IV, V} can be set for each performance index corresponding to each sub-vector in the vector {Q, Q1, Q2, Q3, Q4} representing the service quality (i.e., the score value of each target performance index). According to the boundary corresponding to the preset grading standard, i.e., each level corresponds to a score value range, the boundary is the minimum value and the maximum value of the score value range, three elements of the cloud model, i.e., expectation (E x ), entropy (E n ), and hyper entropy (H e ), are determined, and the cloud model is obtained.

[0053] The index Q i of the evaluation vector is calculated. The expectation Ex i,k at level k is:

[0054]

[0055] The index Q i of the evaluation vector is calculated. The entropy En i,k at level k is:

[0056]

[0057] (wherein and represent the maximum and minimum boundary values of the energy efficiency evaluation index Q i at level k)

[0058] The index Q iHyper-entropy He at level k i,k The constant value is assigned.

[0059] Optionally, each secondary performance indicator is composed of T tertiary performance indicators, T is a positive integer, and the target model is obtained by: acquiring indicator data, wherein the indicator data includes: indicator values of all tertiary performance indicators in a target time period, and score values of the primary performance indicator in the target time period; and training an initial neural network model based on the secondary performance indicators, the indicator values of all tertiary performance indicators in the target time period, and the score values of the primary performance indicator in the target time period, to obtain the target model.

[0060] In this embodiment, the weights of the performance indicators at different levels can be determined by training an artificial neural network, and a supervised learning method can be used in the model training process. Specifically, the secondary performance indicators (for example, the four sub-indicators of availability, throughput, latency, and benefit) can be used as the hidden layer of the neural network, the respective feature indicators (corresponding to the tertiary performance indicators) of the secondary performance indicators can be used as the input of the input layer, and the learning goal of the final neural network is the data center service quality Q (the score value of the primary performance indicator). The neural network is constructed based on the service quality evaluation system.

[0061] In the process of training the initial neural network model, the user score (the score value of the primary performance indicator) can be used as the learning standard. In this embodiment, the service quality provided by the data center can be described by the vector {Q, Q1, Q2, Q3, Q4}, and the score values of the user are the scores of the total service quality, availability, throughput, latency, and benefit, that is, the corresponding vector {Q, Q1, Q2, Q3, Q4}. Then, the collected data of the feature indicators are input to the input layer and the neural network to train the initial neural network model and obtain the target model. This achieves the technical effect of improving the effectiveness of the training target model.

[0062] Optionally, the step of training the initial neural network model based on the secondary performance indicators, the indicator values of all tertiary performance indicators in the target time period, and the score values of the primary performance indicator in the target time period to obtain the target model includes: determining an activation function of the hidden layer of the initial neural network model based on the S secondary performance indicators; inputting the indicator values of all tertiary performance indicators in the target time period as the input of the input layer of the initial neural network, and inputting the score values of the primary performance indicator as the training label; and training the initial neural network model based on the activation function to obtain the target model.

[0063] Figure 3 is a schematic diagram of an optional neural network according to an embodiment of the application, as Figure 3The Leaky RELU (activation function) corresponding to the availability Q1, the throughput Q2, the latency Q3, and the benefit Q4 can be used as the hidden layer of the neural network, and the score value of the first-level performance indicator (e.g., the quality of service Q of the data center in the first-level performance indicator) can be used as the training label. Figure 3 The X11, X12, X13, X21, X22, X23, X31, X32, and X33 representing the third-level performance indicators can be used as the input of the input layer of the initial neural network model, and the initial neural network model is trained. Figure 3

[0064] Optionally, based on the target model, the step of determining the weight of each second-level performance indicator comprises: obtaining the weight coefficient associated with the target model; and determining the weight of each second-level performance indicator based on the weight coefficient.

[0065] In this embodiment, the weight coefficient associated with the target model can be obtained, and the weight of each second-level performance indicator can be determined. For example, in this embodiment, the second-level performance indicators (the four sub-indicators of availability, throughput, latency, and benefit) are associated with the hidden layer of the neural network, and therefore, after the target model is obtained by training the initial neural network model, the weight of each second-level performance indicator can be determined based on the weight coefficient associated with the target model, thereby achieving the technical effect of improving the accuracy of the weight of the second-level performance indicator.

[0066] Optionally, the step of determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each quality of service level comprises: determining the maximum membership in the membership of the to-be-evaluated object belonging to each quality of service level; and determining the evaluation result based on the quality of service level associated with the maximum membership.

[0067] In this embodiment, the quality of service level associated with the maximum membership in the membership of the to-be-evaluated object belonging to each quality of service level can be used as the quality of service level of the to-be-evaluated object, i.e., the evaluation result is obtained.

[0068] In an optional manner, the algorithm for calculating the maximum membership is as follows:

[0069] Input: evaluation vector of the quality of service of the data center (corresponding to the score values of the N target performance indicators), and preset grading standard of each performance indicator;

[0070] Output: maximum membership U p,k , credibility factor θ;

[0071] ​(1) Initialize parameter Er x (Variable) and temp (temporary vector) are both 0;

[0072] (2) for i = 1:h, where h represents the number of performance indicators;

[0073] (3) for each evaluation vector do;

[0074] (4) for each indicator do;

[0075] (5) Calculate the index Q of the evaluation vector. i Expectation at level k: Ex i,k :

[0076]

[0077] (6) Calculate the index Q of the evaluation vector. i Entropy En at level k i,k :

[0078]

[0079] (in and These represent the energy efficiency assessment indicators Q. i (Maximum and minimum boundary values ​​at level k)

[0080] (7) is the index Q of the evaluation vector. i The hyperentropy He at level k i,k Assign a constant value;

[0081] (8) Generate a variable En′ that follows a normal distribution. i,k ;

[0082] (9) Calculate the membership degree μ(x) i,k ):

[0083]

[0084] (10)endfor;

[0085] (11)endfor;

[0086] (12) Calculate the comprehensive evaluation score r(i), where b k Let j be the score for level j, and the corresponding assessment level scores be {1,2,...,k}:

[0087]

[0088]

[0089] (13)Erx = Er x + r(i), update parameter Er x ;

[0090] (14) endfor;

[0091] (15) Er x = h * Er x = mean of calculation result;

[0092] (16) calculate entropy Er h of comprehensive evaluation score

[0093]

[0094]

[0095] (17) calculate credibility factor θ = Er h / Er x , when the credibility factor meets the requirements, proceed to the next step

[0096] (18) output maximum membership degree U p,k , θ:

[0097] K * = max{U p,1 , U p,2 , …, U p,k}

[0098] Through the maximum membership degree, the service quality level of the to-be-evaluated object can be determined, and the technical effect of improving the accuracy of determining the service quality level of the to-be-evaluated object is realized.

[0099] Optionally, after determining the evaluation result of the to-be-evaluated object based on the membership degree of the to-be-evaluated object belonging to each service quality level, the method further comprises: generating a cloud chart associated with the cloud model, wherein the cloud chart is used at least to display the mean value of each target performance indicator; and displaying the cloud chart.

[0100] In the embodiment, after determining the service quality level of the to-be-evaluated object through the membership degree, the service level of the data center under each performance indicator can also be displayed in the cloud chart of the cloud model. In the embodiment, based on the evaluation method of the cloud model, the service quality level corresponding to each data center can be obtained, the quantitative evaluation result is converted into qualitative classification language expression, and it is avoided that the data center is extremely complex and it is difficult to make extremely accurate requirements for the energy efficiency evaluation of the data center. Through the classification, it is more conducive to the planning and supervision of the data center.

[0101] Optionally, after determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each service quality level, the method further comprises: determining a first sequence based on the membership of the to-be-evaluated object belonging to each service quality level, wherein the type of the first sequence comprises a mother sequence associated by a grey correlation analysis method; determining a second sequence based on the N target performance indicators, wherein the type of the second sequence comprises a characteristic sequence associated by the grey correlation analysis method; and determining the correlation degree of each target performance indicator and the evaluation result by the grey correlation analysis method based on the first sequence and the second sequence.

[0102] In this embodiment, based on the membership obtained by the cloud model, the factors (performance indicators) affecting the service quality of the data center can be analyzed by the grey correlation analysis method. In the grey correlation analysis, the membership can be taken as the mother sequence and the indicators can be taken as the characteristic value sequence, so that the correlation degree of each performance indicator to the evaluation result can be obtained, which facilitates the targeted improvement of the service quality of each data center.

[0103] Figure 4 is a schematic diagram of an optional data center service quality evaluation model according to an embodiment of the present application, as shown in Figure 4 In this embodiment, the score value of the user score of the data center service is taken as the guidance and the model training target, an artificial neural network is constructed based on the service quality evaluation system, the weights of the performance indicators are obtained by reverse training, the grading standards of the indicators in the service quality evaluation system are determined, then the characteristic value expectation (Exi), the entropy (Eni) and the hyper entropy (Hei) of the cloud model corresponding to each indicator are calculated, the cloud evaluation model corresponding to each performance indicator (such as the cloud evaluation model 1, the cloud evaluation model 2, the cloud evaluation model 3 and the cloud evaluation model 4 in Figure 4 ), then the membership is calculated based on the obtained weights, the service quality evaluation level is obtained according to the membership, the quantitative analysis is converted into the qualitative analysis by the cloud model, the grading of the service quality of the data center is realized, and the elements affecting the service quality are analyzed by the grey correlation analysis method, which facilitates the improvement of the service quality of the data center.

[0104] Through the embodiment, the service quality of the data center is evaluated based on the cloud model, the previous quantitative evaluation mode is changed, the problem that the complexity of each data center itself cannot be considered in the related art, and the service quality evaluation result deviates from the characteristics of the data center itself is solved, the embodiment considers the characteristics that the evaluation task caused by the complexity of the data center has fuzziness, the quantitative evaluation result is converted to the qualitative language description based on the cloud model, that is, the rating of the service quality, the service quality rating of the data center is performed through the embodiment, which is more in line with the actual quality evaluation task of the data center, and is beneficial to the planning and supervision of the data center industry. And in the quantitative evaluation stage, the user evaluation is oriented, the index weight of each data center is determined, so that the result of the quantitative evaluation stage is more convincing. In addition, after the service quality of the data center is rated, the various elements affecting the service quality are analyzed by the grey correlation analysis method, which provides a reference for improving the service quality of the data center and improves the efficiency of improving the service quality of the data center.

[0105] Embodiment two

[0106] The embodiment two of the application provides an optional data center service quality evaluation device, and each implementation unit in the evaluation device corresponds to each implementation step in the embodiment one.

[0107] Figure 5 It is a schematic diagram of an optional data center service quality evaluation device according to the embodiment of the application, as shown in Figure 5 The evaluation device comprises a first acquisition unit 51, a first determination unit 52, a second determination unit 53 and a third determination unit 54.

[0108] Specifically, the first acquisition unit 51 is configured to acquire score values of N target performance indicators of a to-be-evaluated object, wherein the to-be-evaluated object comprises a data center with a to-be-evaluated service quality rating, and the N target performance indicators comprise a primary performance indicator and S secondary performance indicators, the primary performance indicator is composed of the S secondary performance indicators, N and S are positive integers greater than 1, and S is less than N.

[0109] The first determination unit 52 is configured to determine a cloud model based on a preset grading standard and the score values of each target performance indicator, wherein the preset grading standard is used to evaluate the service quality rating of the to-be-evaluated object.

[0110] The second determination unit 53 is configured to determine the weight of each secondary performance indicator based on a target model, wherein the target model is a neural network model obtained by training an initial neural network model based on index data in a target time period.

[0111] The third determination unit 54 is configured to determine the membership degree of the to-be-evaluated object belonging to each service quality level based on the cloud model and the weight of each secondary performance indicator, and determine the evaluation result of the to-be-evaluated object based on the membership degree of the to-be-evaluated object belonging to each service quality level, wherein the evaluation result is used to indicate the service quality level of the to-be-evaluated object.

[0112] In the data center service quality evaluation apparatus provided in Embodiment Two of the present application, the first acquisition unit 5 is configured to acquire the score value of the N target performance indicators of the to-be-evaluated object, wherein the to-be-evaluated object includes a data center with a to-be-evaluated service quality level, and the N target performance indicators include primary performance indicators and S secondary performance indicators, the primary performance indicators are composed of the S secondary performance indicators, N and S are positive integers greater than 1, and S is less than N. The first determination unit 52 is configured to determine a cloud model based on a preset classification standard and the score value of each target performance indicator, wherein the preset classification standard is used to evaluate the service quality level of the to-be-evaluated object. The second determination unit 53 is configured to determine the weight of each secondary performance indicator based on a target model, wherein the target model is a neural network model obtained by training an initial neural network model based on the indicator data in a target time period. The third determination unit 54 is configured to determine the membership degree of the to-be-evaluated object belonging to each service quality level based on the cloud model and the weight of each secondary performance indicator, and determine the evaluation result of the to-be-evaluated object based on the membership degree of the to-be-evaluated object belonging to each service quality level, wherein the evaluation result is used to indicate the service quality level of the to-be-evaluated object. Thus, the technical problem of poor evaluation effect caused by the fact that the related art quantitatively evaluates the service quality of the data center by taking the total evaluation indicator as the criterion is solved. In this embodiment, the weight of the performance indicator is determined by the neural network model, the service quality level of the to-be-evaluated object is determined by the cloud model based on the weight and the score value of the performance indicator, the situation that the related art ignores the characteristics of the data center and evaluates the service quality of the data center by uniformly weighting all indicators is avoided, and the accuracy of the evaluation of the service quality of the data center is improved.

[0113] Optionally, in the data center service quality evaluation apparatus provided in Embodiment Two of the present application, the first determination unit includes: a first determination subunit configured to determine the boundary value corresponding to each service quality level based on the preset classification standard, wherein the boundary value includes the minimum value and the maximum value of the score value range associated with each service quality level; a second determination subunit configured to determine the digital feature of the cloud model based on the boundary value corresponding to each service quality level, wherein the digital feature includes the expectation, the entropy, and the hyper entropy; and a third determination subunit configured to determine the cloud model based on the digital feature.

[0114] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, each secondary performance indicator is composed of T tertiary performance indicators, T is a positive integer, and the target model is obtained through the following units: a second acquisition unit, configured to acquire indicator data, wherein the indicator data includes: indicator values of all tertiary performance indicators in a target time period, and score values of the primary performance indicator in the target time period; and a training unit, configured to train an initial neural network model based on the secondary performance indicators, the indicator values of all tertiary performance indicators in the target time period, and the score values of the primary performance indicator in the target time period, to obtain the target model.

[0115] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, the training unit includes: a fourth determination subunit, configured to determine an activation function of a hidden layer of the initial neural network model based on the S secondary performance indicators; and a training subunit, configured to take the indicator values of all tertiary performance indicators in the target time period as inputs of an input layer of the initial neural network, take the score values of the primary performance indicator as training labels, and train the initial neural network model based on the activation function to obtain the target model.

[0116] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, the second determination unit includes: an acquisition subunit, configured to acquire a weight coefficient associated with the target model; and a fifth determination subunit, configured to determine a weight of each secondary performance indicator based on the weight coefficient.

[0117] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, in the data center service quality evaluation device provided in Embodiment Two of the present application, the third determination unit includes: a sixth determination subunit, configured to determine a maximum membership degree among the membership degrees of the to-be-evaluated object belonging to each service quality level; and a seventh determination subunit, configured to determine the evaluation result based on a service quality level associated with the maximum membership degree.

[0118] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, the data center service quality evaluation device further includes: a generation unit, configured to generate a cloud chart associated with the cloud model after determining the evaluation result of the to-be-evaluated object based on the membership degrees of the to-be-evaluated object belonging to each service quality level, wherein the cloud chart is used to at least display a mean value of each target performance indicator; and a display unit, configured to display the cloud chart.

[0119] Optionally, in the data center service quality evaluation device provided in Embodiment Two of the present application, the data center service quality evaluation device further comprises a fourth determination unit configured to, after determining the evaluation result of the to-be-evaluated object based on the membership of the to-be-evaluated object belonging to each service quality level, the method further comprises: a fifth determination unit configured to determine a first sequence based on the membership of the to-be-evaluated object belonging to each service quality level, wherein the type of the first sequence comprises a mother sequence associated by a grey correlation analysis method; a sixth determination unit configured to determine a second sequence based on the N target performance indicators, wherein the type of the second sequence comprises a characteristic sequence associated by the grey correlation analysis method; and a seventh determination unit configured to determine the correlation degree of each target performance indicator and the evaluation result based on the first sequence and the second sequence by the grey correlation analysis method.

[0120] The data center service quality evaluation device described above can further comprise a processor and a memory, and the first acquisition unit 51, the first determination unit 52, the second determination unit 53, the third determination unit 54, and the like described above are stored in the memory as program units, and the corresponding functions are realized by the processor executing the program units stored in the memory.

[0121] The processor described above comprises a kernel, and the corresponding program units are called from the memory by the kernel. The kernel can be one or more, and the weight of the performance indicator is determined by adjusting the kernel parameters through the neural network model. Based on the weight and the score value of the performance indicator, the service quality level of the to-be-evaluated object is determined through the cloud model, thereby avoiding the situation that the related art ignores the characteristics of the data center itself, uniformly weights all indicators to evaluate the service quality of the data center, and the quality evaluation result is not accurate, thereby realizing the technical effect of improving the accuracy of data center service quality evaluation.

[0122] The memory described above can comprise a non-permanent memory in a computer readable medium, a random access memory (RAM), and / or a non-volatile memory such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory comprises at least one memory chip.

[0123] According to another aspect of the embodiments of the present application, an electronic device is further provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the data center service quality evaluation method of any one of the above by executing the executable instructions.

[0124] According to another aspect of the embodiments of the present application, a computer readable storage medium is further provided, which stores a computer program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute the data center service quality evaluation method of any one of the above when the computer program runs.

[0125] Figure 6 is a schematic diagram of an electronic device according to an embodiment of the present application, as Figure 6 shown, the embodiment of the present application provides an electronic device 60, which comprises a processor, a memory, and a program stored in the memory and capable of running on the processor, and the processor implements the data center service quality evaluation method of any one of the above when executing the program.

[0126] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0127] In the above-mentioned embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0128] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other ways. Among them, the above-mentioned device embodiments are only schematic, for example, the division of the units can be a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0129] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed to multiple units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0130] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0131] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0132] The above is only the preferred embodiment of the present application, it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, can make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for evaluating the quality of data center services, characterized in that, include: Obtain the score values ​​of N target performance indicators of the object to be evaluated, wherein the object to be evaluated includes: the data center whose service quality level is to be evaluated, and the N target performance indicators include: primary performance indicators and S secondary performance indicators. The primary performance indicators are composed of the S secondary performance indicators, and N and S are positive integers greater than 1, and S is less than N. Based on the preset grading standards and the score value of each target performance indicator, a cloud model is determined, wherein the preset grading standards are used to evaluate the service quality level of the object to be evaluated. Based on the target model, the weight of each of the secondary performance indicators is determined, wherein the target model is a neural network model obtained by training an initial neural network model based on indicator data within the target time period; Based on the cloud model and the weight of each of the secondary performance indicators, the membership degree of the object to be evaluated to each service quality level is determined, and the evaluation result of the object to be evaluated is determined based on the membership degree of the object to be evaluated to each service quality level, wherein the evaluation result is used to indicate the service quality level of the object to be evaluated. Each of the secondary performance indicators consists of T tertiary performance indicators, where T is a positive integer. The target model is obtained in the following way: The indicator data is obtained, wherein the indicator data includes: the indicator values ​​of all third-level performance indicators within the target time period, and the score values ​​of the first-level performance indicators within the target time period; Based on the secondary performance indicators, the index values ​​of all tertiary performance indicators within the target time period, and the score values ​​of the primary performance indicators within the target time period, the initial neural network model is trained to obtain the target model. The steps for training the initial neural network model to obtain the target model based on the secondary performance indicators, the indicator values ​​of all tertiary performance indicators within the target time period, and the score values ​​of the primary performance indicators within the target time period include: Based on the S secondary performance indicators, determine the activation function of the hidden layer of the initial neural network model; The values ​​of all third-level performance indicators within the target time period are used as the input of the initial neural network input layer, and the scores of the first-level performance indicators are used as training labels. Based on the activation function, the initial neural network model is trained to obtain the target model. The third-level performance indicators are obtained through the following steps: four sub-indicators are selected as the synthesis of the first-level performance indicators, namely availability, throughput, latency, and efficiency. Several real-time operation and maintenance data are selected from each of the four sub-indicators as the third-level performance indicators.

2. The evaluation method according to claim 1, characterized in that, The steps for determining the cloud model based on preset grading standards and the score value of each target performance indicator include: Based on the preset grading standard, the boundary values ​​corresponding to each service quality level are determined, wherein the boundary values ​​include: the minimum and maximum values ​​of the scoring value range associated with each service quality level; Based on the boundary values ​​corresponding to each service quality level, the digital features of the cloud model are determined, wherein the digital features include: expectation, entropy, and hyperentropy; The cloud model is determined based on the digital characteristics.

3. The evaluation method according to claim 1, characterized in that, The step of determining the weight of each of the secondary performance indicators based on the target model includes: Obtain the weight coefficients associated with the target model; Based on the weighting coefficients, the weight of each of the secondary performance indicators is determined.

4. The evaluation method according to claim 1, characterized in that, The step of determining the evaluation result of the object to be evaluated based on its membership degree to each service quality level includes: Determine the maximum membership degree of the object to be evaluated within each service quality level; The evaluation result is determined based on the service quality level associated with the maximum membership degree.

5. The evaluation method according to claim 1, characterized in that, After determining the evaluation result of the object to be evaluated based on its membership degree to each service quality level, the method further includes: Generate a cloud map associated with the cloud model, wherein the cloud map is used to display at least the mean of each of the target performance metrics; The cloud map is then displayed.

6. The evaluation method according to claim 1, characterized in that, After determining the evaluation result of the object to be evaluated based on its membership degree to each service quality level, the method further includes: Based on the membership degree of the object to be evaluated to each service quality level, a first sequence is determined, wherein the type of the first sequence includes: the parent sequence associated by grey relational analysis; Based on N target performance indicators, a second sequence is determined, wherein the type of the second sequence includes: the feature sequence associated by the grey relational analysis method; Based on the first sequence and the second sequence, the correlation between each target performance index and the evaluation result is determined using the grey relational analysis method.

7. A device for evaluating the quality of service in a data center, characterized in that, include: The acquisition unit is used to acquire the score values ​​of N target performance indicators of the object to be evaluated. The object to be evaluated includes: a data center whose service quality level is to be evaluated. The N target performance indicators include: a primary performance indicator and S secondary performance indicators. The primary performance indicator is composed of the S secondary performance indicators. N and S are positive integers greater than 1, and S is less than N. The first determining unit is used to determine the cloud model based on a preset grading standard and the score value of each of the target performance indicators, wherein the preset grading standard is used to evaluate the service quality level of the object to be evaluated. The second determining unit is used to determine the weight of each of the secondary performance indicators based on the target model, wherein the target model is a neural network model obtained by training an initial neural network model based on indicator data within a target time period; The third determining unit is used to determine the membership degree of the object to be evaluated to each service quality level based on the cloud model and the weight of each of the secondary performance indicators, and to determine the evaluation result of the object to be evaluated based on the membership degree of the object to be evaluated to each service quality level, wherein the evaluation result is used to indicate the service quality level of the object to be evaluated. Each secondary performance indicator consists of T tertiary performance indicators, where T is a positive integer. The target model is obtained through the following units: a second acquisition unit, used to acquire indicator data, which includes: the indicator values ​​of all tertiary performance indicators within the target time period and the score values ​​of the primary performance indicators within the target time period; and a training unit, used to train the initial neural network model based on the secondary performance indicators, the indicator values ​​of all tertiary performance indicators within the target time period, and the score values ​​of the primary performance indicators within the target time period to obtain the target model. The training unit includes: a fourth determining subunit, used to determine the activation function of the hidden layer of the initial neural network model based on S secondary performance indicators; and a training subunit, used to use the index values ​​of all tertiary performance indicators within the target time period as the input of the initial neural network input layer, and the score values ​​of the primary performance indicators as training labels, and to train the initial neural network model based on the activation function to obtain the target model. The third-level performance indicators are obtained through the following steps: four sub-indicators are selected as the synthesis of the first-level performance indicators, namely availability, throughput, latency, and efficiency. Several real-time operation and maintenance data are selected from each of the four sub-indicators as the third-level performance indicators.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the data center service quality assessment method according to any one of claims 1 to 6.

9. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the data center service quality assessment method according to any one of claims 1 to 6.

Images